 |
Wednesday, May 28, 2008 - 1:30 PM
453
Adaptive cluster, two-stage sequential, and two-phase sampling designs for estimation of freshwater mussel population density
David R. Smith1, Rita F. Villella1, Daniel J. Hornbach2, and Mark C. Hove2. (1) Leetown Science Center, Aquatic Ecology Lab, USGS, 11649 Leetown Road, Kearneysville, WV 25430, (2) Department of Biology, Macalester College, 1600 Grand Avenue, St. Paul, MN 55105
Adaptive cluster, two-stage sequential, and two-phase sampling designs are all ‘adaptive’ in the sense that final sampling effort depends on what is found while sampling. In adaptive cluster sampling, additional quadrats are sampled in the neighborhood of quadrats where a rare species is found. In two-stage sequential sampling, an additional sequence of quadrats is selected within primary units where a species is found. In two-phase sampling, qualitative sampling determines how many sites are quantitatively sampled along a river or within a watershed. When applied to rare and clustered populations more rare species are encountered and more individuals from rare species are observed in adaptive sampling designs than in conventional sampling designs. We present each design and discuss guidelines for effective application. We also cite real applications of the designs for estimating density of rare or endangered freshwater mussels in Atlantic-slope rivers (adaptive cluster sampling), the St. Croix River in MN (two-stage sequential sampling), and the Allegheny River in PA (two-phase sampling).
Web Page: adaptive sampling, rare and clustered populations, quantitative sampling